Kakao to Triple Investment in Service Stabilization... Dedicated Team Directly Under CEO

Full Multi-Redundancy of Service Operation System... Data Center with Triple Redundancy + α
Aggressive Hiring of Engineering Personnel... Global Technology R&D for Incident Response

[Asia Economy Reporter Yuri Choi] Kakao, which experienced a service outage due to a data center fire, will triple its investment in service stabilization to prevent recurrence. The dedicated IT engineering organization will be expanded and elevated directly under CEO Hong Eun-taek, and Go Woo-chan, Chief Cloud Officer (COO) of Kakao Enterprise, will also be brought to headquarters.

On the 7th, Kakao held its annual developer conference, 'If Kakao Dev 2022' (hereafter 'If Kakao'), and announced measures to prevent recurrence. This is a plan released about two months after the outage caused by the fire at SK C&C Pangyo Data Center on October 15.

First, investment in service stabilization will be expanded. Over the next five years, more than three times the amount invested in the past five years in infrastructure establishment, server-related equipment purchases, and information security investment will be allocated. However, the specific investment amount was not disclosed due to trade secrets.

Specifically, the entire service operation system will be duplicated. The main backbone data centers hosting key services will be increased from the current two to three, and facility investments will be made to handle the increased traffic between data centers. For services requiring large-capacity traffic transmission, a separate dedicated network will be configured for triplication. All types of data will be multiply replicated, and operation management tools will also be triplicated. Even if a failure occurs throughout the data center, the monitoring system will be duplicated to ensure monitoring and fault detection work properly.

The Ansan Data Center, scheduled to operate in 2024, will strengthen its disaster prevention functions. The uninterruptible power supply (UPS) room and battery room, identified as one of the causes of the Pangyo Data Center fire, will be separated by fireproof partitions. This is to prevent a fire in the battery room from affecting the operation of other facilities. In case of a fire in the battery room, a triple extinguishing method will be applied.

An emergency response plan and data center disaster recovery (DR) system will also be established to prepare for unexpected situations. The DR structure will be triplicated so that even if one of the three data centers is disabled, redundancy is guaranteed. Critical services, such as KakaoTalk message transmission, will add external cloud as a safety measure. In emergencies, safety measures will be prepared up to triplication plus alpha (α).

Go, COO and co-chairman of Kakao's Emergency Countermeasures Committee for recurrence prevention, explained, "Each recurrence prevention measure requires bold investment decisions amounting to hundreds of billions or even trillions of won annually in operating costs, excluding initial investment."

The dedicated service stabilization organization will also be expanded. The IT engineering organization, currently under the development organization, will be placed directly under the CEO and its scale increased. Top experts in various fields such as data centers and cloud development engineers will be actively recruited. Additionally, a disaster recovery committee to prepare for large-scale failures and a dedicated organization prioritizing service continuity will be newly established.

In the R&D area, Kakao will develop technologies not yet widespread domestically but adopted by major global companies. They are reviewing chaos engineering, which intentionally induces failures to test system stability, and service status pages that monitor service conditions.

Namgoong Hoon, former Kakao CEO and co-chairman of the Emergency Countermeasures Committee for recurrence prevention, emphasized, "We will reflect and improve even now to do our best to prevent such accidents in the future," adding, "Service stabilization is the top priority and a social responsibility that we will always keep in mind."

Meanwhile, on the second day of If Kakao, the 8th, Kakao will unveil five special sessions titled '1015 Retrospective.' These will individually explain the duplication technologies applied in five areas: data centers, infrastructure facilities, data, service platforms, and applications.